---
title: Wiki Index
tags:
  - wiki
  - index
date: "2026-07-17T06:13:45.940Z"
created: "2026-07-17T06:13:45.940Z"
includeInRss: false
---

# Wiki Index

Catalog of agent-maintained wiki pages. Updated by `/ingest`. Prefer this before deep QMD for wiki-scoped questions.

## Overview

| Page | Summary |
| --- | --- |
| [[README\|Wiki]] | Layer map, ownership, operations |
| [[Vault architecture synthesis]] | How this vault maps capture → raw → wiki |
| [[Agentic stack synthesis]] | IDE agents, MCP, personal agents, wiki |

## Sources (64)

| Page | Summary | Raw |
| --- | --- | --- |
| [[A Field Guide to Fable: Finding Your Unknowns]] | Prompts/skills/context are the **map**; the codebase and real constraints are the **territory**—gaps between them are ** | [[A Field Guide to Fable: Finding Your Unknowns]] |
| [[A sane but extremely bull case on OpenClaw (Clawdbot)]] | OpenClaw hype (unlimited permissions, token burn, agent social networks) obscures a real shift: persistent personal agen | [[A sane but extremely bull case on OpenClaw (Clawdbot)]] |
| [[AGENTS.md outperforms skills in our agent evals]] | Vercel evals on Next.js 16 APIs: an ~8KB compressed docs index in `AGENTS.md` hit **100%** pass rate; skills capped at * | [[AGENTS.md outperforms skills in our agent evals]] |
| [[AI-Augmented Software Development Manifesto]] | Values human agency/taste, tight feedback loops, reliable tested software, and adaptable intent over autopilot, long una | [[AI-Augmented Software Development Manifesto]] |
| [[AI-Native Obsidian Vault Setup Guide]] | Structures vault as a **life OS**: numeric folder prefixes, periodics (daily/weekly/monthly/quarterly), entities (people | [[AI-Native Obsidian Vault Setup Guide]] |
| [[Advanced React]] | Wes Bos course landing page for **Fullstack Advanced React & GraphQL**—build "Sick Fits" e-commerce (search, cart, check | [[Advanced React]] |
| [[Async Ruby is the Future of AI Apps (And It's Already Here)]] | LLM apps need long-lived connections, token streaming, and many concurrent mostly-idle sessions—thread-based job queues  | [[Async Ruby is the Future of AI Apps (And It's Already Here)]] |
| [[Beyond the Prompt: Claude Code]] | Core Boris principle: **give Claude a verification loop** (tests, linters, runnable checks)—claimed 2–3× quality vs. hum | [[Beyond the Prompt: Claude Code]] |
| [[Bitcoin A Peer-to-Peer Electronic Cash System]] | Satoshi whitepaper (Genius annotated edition): peer-to-peer electronic cash without financial institutions as trusted th | [[Bitcoin A Peer-to-Peer Electronic Cash System]] |
| [[Build Your First AI Agent in Just 5 Minutes!]] | Thesis: agents are commoditizing; **marketing** is an early mass-adoption surface (LinkedIn content, social monitoring,  | [[Build Your First AI Agent in Just 5 Minutes!]] |
| [[Building Your AI-Powered Second Brain Claude Code and Obsidian]] | Problem frame: knowledge scattered across apps; cognitive overhead of "where did I put that?" exceeds thinking time. | [[Building Your AI-Powered Second Brain Claude Code and Obsidian]] |
| [[Building efficient MCP servers]] | MCP standardizes AI model integrations; Vercel's open-source MCP adapter targets Next.js/Nuxt/SvelteKit deploys (Zapier, | [[Building efficient MCP servers]] |
| [[Claude Code Skills for Solopreneurs]] | **opc.dev** bundles nine free open-source skills: SEO/GEO, RequestHunt demand research, domain hunter, NanoBanana image  | [[Claude Code Skills for Solopreneurs]] |
| [[Cursor agent best practices]] | Agent harness = **instructions** (rules/prompts) + **tools** (edit, search, terminal) + **model**; Cursor tunes per-mode | [[Cursor agent best practices]] |
| [[Five foundations for building complex Rails apps]] | Vanilla Rails suffices until ~100–150k LOC; beyond that, maintainability needs new technical *and* mental foundations. | [[Five foundations for building complex Rails apps]] |
| [[Fragments January 8]] | Anthropic internal report: AI used for **59%** of dev work, ~**50%** productivity lift; debugging and understanding code | [[Fragments January 8]] |
| [[Getting started with loops]] | Claude Code defines **loops** as agents repeating work cycles until a stop condition is met — not every task needs a com | [[Getting started with loops]] |
| [[Here's how I use LLMs to help me write code]] | LLM-assisted coding is **difficult and unintuitive** — treat models as an over-confident, fast pair programmer, not AGI; | [[Here's how I use LLMs to help me write code]] |
| [[How I Use Claude Code]] | Core rule: **never let Claude write code until a written plan is reviewed and approved** — separates architecture judgme | [[How I Use Claude Code]] |
| [[How I use Obsidian — Steph Ango]] | Vault = folder of files (**file over app**); bottom-up, chaos-tolerant organization that lets structure emerge rather th | [[How I use Obsidian — Steph Ango]] |
| [[How to enter side doors]] | The job market is a **search problem**, not a front-door application funnel — careers often start via side gates, warm i | [[How to enter side doors]] |
| [[How to implement Stripe without going mad]] | Stripe's core pain is **split brain**: purchase state lives in Stripe while your app tracks via webhooks — 258+ event ty | [[How to implement Stripe without going mad]] |
| [[I turned my Obsidian vault into an AI-powered work OS — open source, free]] | **ObsidianOS** (`benoror/obsidianos_work`) is an open-source (MIT) Obsidian vault with agent skills run via slash comman | [[I turned my Obsidian vault into an AI-powered work OS — open source, free]] |
| [[Introducing Eve]] | **Eve** (Vercel, open source) is an agent framework with production primitives built in: durable execution, sandboxed co | [[Introducing Eve]] |
| [[Introducing Moltworker: a self-hosted personal AI agent, minus the minis]] | **Moltworker** (Cloudflare) runs **OpenClaw** (formerly Moltbot/Clawdbot) on Workers + Sandbox SDK — self-hosted persona | [[Introducing Moltworker: a self-hosted personal AI agent, minus the minis]] |
| [[Introducing the Model Context Protocol]] | **MCP** (Anthropic, open standard) connects AI assistants to data sources — repos, business tools, dev environments — vi | [[Introducing the Model Context Protocol]] |
| [[Karpathy LLM Wiki]] | **LLM Wiki pattern**: instead of RAG-only retrieval, the LLM incrementally builds and maintains a **persistent, interlin | [[Karpathy LLM Wiki]] |
| [[Managing remote teams a crash course]] | Remote setups differ: **satellite offices**, **remote employees** (minority remote), **fully distributed**, and **remote | [[Managing remote teams a crash course]] |
| [[Moltbook is the most interesting place on the internet right now]] | **OpenClaw** (Clawdbot → Moltbot → OpenClaw) is a fast-growing open-source personal assistant integrating with messaging | [[Moltbook is the most interesting place on the internet right now]] |
| [[My AGENTS.md file for building plans you actually read]] | **Plan loop**: Plan → Execute → Test → Commit — skipping planning forces the AI to guess and produces hallucinations and | [[My AGENTS.md file for building plans you actually read]] |
| [[News Scan — 2026-07-12]] | Initial `/news-scan` **backlog reset**: 17,941 newsletter messages (2021–2026) → archived 17,695 (>30d backlog + recent  | [[News Scan — 2026-07-12]] |
| [[Newsletter Digest — 2026-07-12]] | **AI/agents theme**: model economics shifting (Grok 4.5 in Cursor), **Assisted → Delegated → Autonomous** maturity ladde | [[Newsletter Digest — 2026-07-12]] |
| [[Night Shift Agentic Workflow]] | **Day shift / night shift split:** Human time is for specs, architecture, and thinking; agents run autonomously overnigh | — |
| [[Notion Tools and Craft Ted Nelson]] | Notion *Tools and Craft* series entry on **Ted Nelson** (hypertext pioneer, Xanadu). | — |
| [[On the Biology of a Large Language Model]] | Anthropic **interpretability** paper applying a **biology metaphor** to LLMs: simple training algorithms produce complex | — |
| [[Open in Safari [Shortcuts]]] | **Problem:** Arc (macOS default) + Arc Search (iOS) lack Safari-only shortcut actions and integrations. | — |
| [[OpenClaw Our Comprehensive Guide for Beginners]] | **Every.to "Claw School"** onboarding for **OpenClaw** — Peter Steinberger's open-source personal AI assistant framework | — |
| [[OpenClaw Security 101 The Complete Guide]] | OpenClaw can run commands, read files, send messages, and call APIs — **compromise = full account takeover** (keys, prom | — |
| [[OpenClaw, OpenAI and the future]] | **Peter Steinberger joins OpenAI** to work on bringing agents to everyone; mission shift: build an agent "even my mum ca | — |
| [[Operational Transformation]] | Topic: **operational transformation (OT)** — algorithms for real-time collaborative editing / conflict resolution. | — |
| [[Professional screen recorder for macOS]] | **Screen Studio** (screen.studio) — macOS screen recorder for product demos, courses, tutorials, and social videos. | — |
| [[Promoting AI agents]] | **DHH's stance shift:** agents "came alive" when given tools beyond reasoning — terminal, tests, web search, **Agent Ski | — |
| [[Rails 6 adds db seed replant task]] | **`rails db:seed:replant`** (Rails 6+): truncates all tables for the **current environment**, then loads `db/seeds.rb`. | — |
| [[Rails Dockerfile Best Practices Part 1 Image Slimming]] | **Series goal:** line-by-line walkthrough of Rails 7.1+ default `Dockerfile` — Part 1 covers **image size reduction** on | — |
| [[Salary Negotiation Make More Money Be More Valued]] | **Patrick McKenzie (kalzumeus):** engineers treat salary negotiation as virtue signaling; **5 minutes of negotiation** r | — |
| [[Secret Sauce Growth Hacking Guide]] | **Secret Sauce** growth-hacking guide at [secretsaucenow.com](https://www.secretsaucenow.com/) — content not captured in | — |
| [[Shipping at Inference-Speed]] | **Peter Steinberger (Dec 2025):** shipping speed now limited by **inference time + hard thinking** — most software is bo | — |
| [[The Omarchy Manual]] | **Omarchy** is DHH's omakase Arch + Hyprland Linux distro—curated dev stack (Neovim, Alacritty, Chromium, Typora, etc.)  | [[The Omarchy Manual]] |
| [[The PARA Method]] | **PARA** (Tiago Forte): four categories for all digital information—**Projects**, **Areas**, **Resources**, **Archives** | [[The PARA Method]] |
| [[The alternative to shared examples]] | Argues against RSpec `shared_examples` for mix-ins in favor of **isolated contract-tests**—spec the mixin module directl | [[The alternative to shared examples]] |
| [[The boring technology behind a one-person Internet company]] | Clipped from [HN item 20985875](https://news.ycombinator.com/item?id=20985875); original article likely at Listen Notes  | [[The boring technology behind a one-person Internet company]] |
| [[The next chapter EpicAI.pro]] | Kent C. Dodds announces **EpicAI.pro**—next education platform after Testing JavaScript, Epic React, and Epic Web—focuse | [[The next chapter EpicAI.pro]] |
| [[They're Made Out of Weights]] | Max Leiter's parody of Terry Bisson's "They're Made Out of Meat"—aliens discover LLMs are **only floating-point weights* | [[They're Made Out of Weights]] |
| [[They're Made out of Meat | East of the Web]] | Terry Bisson's classic short story: alien intelligence discovers humans are **entirely meat**—born meat, die meat, brain | [[They're Made out of Meat | East of the Web]] |
| [[This is a sample AGENTS.md file from my classes]] | Sample **AGENTS.md** for university CS courses: AI assistants are **teaching aids**, not assignment solvers. | [[This is a sample AGENTS.md file from my classes]] |
| [[Twitch live coding LowCode streaming]] | Jordan Lewis's 6-month retrospective on **LARGE DATA BANK** Twitch live-coding channel—setup guide for coding streams as | [[Twitch live coding LowCode streaming]] |
| [[Unit Testing React Components]] | Eric Elliott on **TDD for React**: unit tests reduce bug density, improve architecture/API design, and give fast file-sa | [[Unit Testing React Components]] |
| [[Why we sunsetted mcpt]] | Mintlify built **mcpt.com** (MCP server registry) in one week after viral MCP generator launch; shut down **five days la | [[Why we sunsetted mcpt]] |
| [[Writing about Agentic Engineering Patterns]] | Simon Willison launches **[Agentic Engineering Patterns](https://simonwillison.net/guides/agentic-engineering-patterns/) | [[Writing about Agentic Engineering Patterns]] |
| [[Zen of AI Coding]] | Yoav Aviram's 16-aphorism **Zen of Python**–style manifesto: marginal cost of code is collapsing, so **software developm | [[Zen of AI Coding]] |
| [[a meta-harness for building and running AI agents]] | **Omnigent** (omnigent.ai) is an open-source **meta-harness**: a common layer over Claude Code, Codex, Pi, and custom ag | — |
| [[an intro to mcp for busy devs]] | Allen Thomas TL;DR: **MCP server : LLM :: LSP : text editor**; MCP is "ODBC for AI"—one protocol instead of per-LLM tool | [[an intro to mcp for busy devs]] |
| [[assignment1-basics/CLAUDE.md at main]] | Stanford **CS336** (Language Modeling From Scratch) **CLAUDE.md**: AI agents are TAs, not solution generators for implem | [[assignment1-basics/CLAUDE.md at main]] |
| [[juggernaut on X: early-stage startup advice (image)]] | Viral X post: CTO shared long-form startup advice image on employee day one; author recommends early-stage founders/empl | [[juggernaut on X: \"CTO shared this with me on day 1something everyone should read if they are in an early-stage startup at least once https://t.co/4M7PR3P46O\"]] |

## Entities (3)

| Page | Summary |
| --- | --- |
| [[Andrej Karpathy]] | LLM Wiki pattern → `Wiki/` + `/ingest` |
| [[Claude Code and Cursor agents]] | Plan/loop modes, skills, and verification rituals transfer across hosts. |
| [[OpenClaw]] | Personal persistent AI agent project (aliases: Clawdbot / Moltbot lineage in coverage). Messaging-native, self-extending |

## Concepts (7)

| Page | Summary |
| --- | --- |
| [[AGENTS.md vs skills]] | Persistent agent files beat piles of skills when the eval is “does the agent follow house rules.” |
| [[Agentic coding craft]] | **Context is the craft** — what you put in AGENTS/skills/files beats clever prompts. |
| [[Career side doors and negotiation]] | **Side doors** — make work visible; warm intros beat cold apply; signal in noise. |
| [[LLM Wiki pattern]] | **Raw is immutable** — clippings/sources are curated inputs; the wiki is the living artifact. |
| [[Model Context Protocol]] | MCP separates **client** (Cursor, Claude Desktop) from **server** (Gmail, Raindrop, QMD, filesystem). |
| [[Obsidian as agent IDE]] | **File over app** (kepano) — plain files outlive products. |
| [[Personal AI agents]] | Persistence + tools + identity/memory beats chatbox demos. |

## Syntheses (2)

| Page | Summary |
| --- | --- |
| [[Agentic stack synthesis]] | Session agents for vault surgery; personal agents for always-on ops — don’t conflate. |
| [[Vault architecture synthesis]] | How this personal vault maps intentionality + Karpathy compile layer. |